JP6464891B2 - Sideline problem scoring method, sideline problem scoring program, and sideline problem scoring device - Google Patents

Description

  The present invention relates to a side line problem scoring method, a side line problem scoring program, and a side line problem scoring apparatus.

  Electronic teaching materials using the Internet and tablet terminals are being introduced. In such an electronic teaching material, as an example, a sideline problem may be given in a form in which a sideline is drawn at a position corresponding to an answer in a question sentence.

  If such a sideline problem is provided as an electronic teaching material, if the question sentence is text data described in text format, the sideline can be entered as an answer by specifying the range through a drag operation etc. on the computer, You can score answers. However, when the question sentence is a paper document, the paper sentence of the question sentence is OCR (Optical) in order to associate the correct character string included in the question sentence with the character string of the side line portion input as the answer. It is digitized by using Character Recognition.

Japanese Patent Laid-Open No. 3-164885 JP 2007-48217 A

  However, OCR has its own limitations on the accuracy of character recognition. That is, OCR includes procedures such as layout recognition, line direction estimation, line division, character division, and character recognition. Each of these procedures can cause errors due to various causes. For example, an unexpected special layout, blurring or blurring of characters in a paper document of a problem sentence, dirt on a paper surface, and the like cause errors. As described above, when OCR is used, a correction operation on the result of character recognition is unavoidable, and as a result, it takes time to associate the correct character string of the problem sentence with the character string of the side line part, resulting in scoring efficiency of the side line problem. Will fall.

  In one aspect, an object of the present invention is to provide a side line problem scoring method, a side line problem scoring program, and a side line problem scoring apparatus that can improve scoring efficiency of a side line problem.

  In the side line problem scoring method of one aspect, the computer sets a frame that surrounds part or all of the character string on the image including the character string, calculates a determination range based on the set frame, and is based on handwriting input A handwriting is acquired, and based on the position of the frame and the position of the handwriting input by handwriting input, a process of determining whether or not the handwriting is within the calculated determination range is executed.

  The scoring efficiency of the side line problem can be improved.

FIG. 1 is a diagram illustrating a functional configuration of the side line problem scoring device according to the first embodiment. FIG. 2 is a diagram illustrating an example of detection of black pixel breaks. FIG. 3A is a diagram illustrating a setting example of a rectangle in a question sentence image. FIG. 3B is a diagram illustrating an example of entering an answer to the question sentence image. FIG. 3C is a diagram illustrating an example of a first determination method. FIG. 3D is a diagram illustrating an example of a second determination method. FIG. 4 is a flowchart (1) illustrating the procedure of the marginal line problem scoring process according to the first embodiment. FIG. 5 is a flowchart (2) illustrating the procedure of the marginal line problem scoring process according to the first embodiment. FIG. 6 is a diagram illustrating an application example. FIG. 7A is a diagram illustrating one aspect of an effect relating to an application example. FIG. 7B is a diagram illustrating one aspect of an effect related to the application example. FIG. 8 is a diagram illustrating a hardware configuration example of a computer that executes the sideline problem scoring program according to the first embodiment and the second embodiment.

  Hereinafter, a sideline problem scoring method, a sideline problem scoring program, and a sideline problem scoring apparatus according to the present application will be described with reference to the accompanying drawings. Note that this embodiment does not limit the disclosed technology. Each embodiment can be appropriately combined within a range in which processing contents are not contradictory.

[Configuration of sideline problem scoring system]
FIG. 1 is a diagram illustrating a functional configuration of the side line problem scoring device according to the first embodiment. The sideline problem scoring device 10 shown in FIG. 1 determines the correct or incorrect answer of an answer in which a sideline is input by handwriting for a sideline problem that is given in a form in which a sideline is drawn at a location corresponding to the answer in the question sentence. Thus, the side line problem scoring process for supporting the scoring of the side line problem is executed.

  As one embodiment, the sideline problem scoring apparatus 10 can be implemented by installing a sideline problem scoring program in which the above-mentioned sideline problem scoring process is provided as package software or online software on a desired computer. For example, the side line problem scoring program is installed not only on mobile communication terminals such as smartphones, mobile phones and PHS (Personal Handyphone System) but also on mobile terminal devices including tablet terminals and slate terminals. Thus, the mobile terminal device can function as the side line problem scoring device 10. Here, as an example, the following description will be given by illustrating the case where the side line problem scoring device 10 is implemented as a mobile terminal device, but the side line problem scoring program is installed in a stationary terminal device such as a personal computer. Can also be installed.

  In the following, as an example, the case where the scoring function for the side line problem is implemented on the computer used by the problem creator is illustrated, but the scoring function for the side line problem is not necessarily implemented on the computer used by the problem creator. It is not necessary. In other words, if a function is realized that takes as input a handwriting in which a side line is described via handwriting input on an image including a question sentence, that is, an answer to the side line problem, and outputs a determination result of a correct answer or an incorrect answer of the answer In many cases, it may be implemented as a cloud formed by a computer on a network, for example, a server device, a physical machine, or a group of physical machines.

  Such a side line problem scoring device 10 may include various functional units included in a known computer in addition to the functional units illustrated in FIG. 1. For example, when the sideline problem scoring device 10 is mounted as a tablet terminal, it may further include a motion sensor such as an acceleration sensor or an angular velocity sensor. Further, when the side line problem scoring device 10 is implemented as a mobile communication terminal, it may further include a functional unit such as an antenna or a GPS (Global Positioning System) receiver. In FIG. 1, as an example, a functional unit when the side line problem scoring device 10 is implemented as a tablet terminal is illustrated. However, when the side line problem scoring device 10 is implemented as a stationary terminal, a keyboard is used. In addition, an input / output device such as a mouse or a display may be provided.

  As shown in FIG. 1, the side line problem scoring device 10 includes a touch panel 11, a problem storage unit 12, a setting unit 13, a rectangular storage unit 14, an acquisition unit 15, a handwriting storage unit 16, and a calculation unit 17. And a determination unit 18 and a scoring unit 19.

  The touch panel 11 is a displayable and inputable device.

  As one aspect regarding display, the touch panel 11 displays an image output by an OS (Operating System) or an application program, including a side line problem scoring program executed on the side line problem scoring device 10. As one aspect related to input, the touch panel 11 accepts touch operations such as tap, flick, sweep, pinch-in, and pinch-out performed on the screen of the touch panel 11, including handwritten input using a touch pen or fingers. In addition, although the touch panel 11 was illustrated as an input device which performs the instruction | indication input with respect to the side line problem scoring apparatus 10 here, it has not only this but the physical key etc. which implement | achieve complementary input between the touch panels 11 etc. It does not matter.

  Here, any method can be adopted as the operation principle of the touch panel 11. For example, resistive film method, capacitance method, surface acoustic wave method, infrared method (whether infrared light is emitted from the edge of the panel parallel to the panel surface and the infrared sensor is blocked by a finger etc. at the opposite end) There are various methods, such as a method of sensing the finger touch position by acquiring whether or not, and a method of tracking the finger position with a camera, but any method may be adopted. In addition to the touch panel 11, an electromagnetic induction method using an electronic pen, a method of specifying a pen tip position by reading a special pattern printed on paper or the like with a camera loaded on the pen tip, an acceleration sensor or a gyro sensor A pen input device or a device attached to the finger, such as a method of measuring the movement of the pen tip or the fingertip, may be adopted.

  The problem storage unit 12 is a storage unit that stores an image including a question sentence. Hereinafter, an image including a question sentence may be referred to as a “problem sentence image”.

  As an embodiment, the problem storage unit 12 stores an image obtained by reading a paper document on which a question sentence is printed by a reading device such as a scanner. As described above, the data related to the problem sentence does not necessarily have to be character string data determined by a character code such as text data, and may be image data.

  The setting unit 13 is a processing unit that sets a predetermined frame on the question sentence image.

  As one embodiment, the setting unit 13 displays a question sentence image on the touch panel 11 when any one of the question sentence images stored in the question storage unit 12 is designated. In the state where the question sentence image is displayed on the touch panel 11 in this way, the setting unit 13 accepts a rectangular setting as an example of a frame that encloses the character string of the part of the question sentence that the question creator defines as the correct answer. For example, when a rectangle is set, the setting of the rectangle can be accepted by specifying a range such as a drag operation. In addition, it is also possible to accept the setting of the rectangle by designating the coordinates of each vertex forming the rectangle or by designating a part of the vertex and the width and height of the rectangle. Here, a case where a rectangular region is set is illustrated as an example of the “frame”, but any shape such as a polygon, a circle, or an ellipse can be set as the frame in addition to the rectangle.

  When the rectangle is set in this way, the setting unit 13 can support the setting of the rectangle by using the detection result of the character break detected from the question sentence image. For example, the setting unit 13 calculates a histogram of black pixels for each direction by scanning a color corresponding to a character in the image in the question sentence image displayed on the touch panel 11 in the vertical direction and the horizontal direction. After that, the setting unit 13 detects a portion where the frequency is equal to or less than a predetermined threshold on the black pixel histogram as a black pixel break, and sets the width and height of the rectangle for which the range is designated by the problem creator to black. It can be expanded and contracted according to pixel breaks. Here, a case where a black pixel break is detected is illustrated, but the application range is not limited to a black and white problem sentence image. That is, it is not always necessary to use a black pixel to detect a character break. When a character is printed in another color, the pixel corresponding to the character color may be scanned.

  FIG. 2 is a diagram illustrating an example of detection of black pixel breaks. In FIG. 2, as an example of a character string included in a horizontally written problem sentence image, a character string “: has ...” and a histogram of black pixels in the vertical direction and the horizontal direction are shown. Has been. As illustrated in FIG. 2, the setting unit 13 scans the problem sentence image in the vertical direction and the horizontal direction around the rectangle for which the range is specified via the touch panel 11, thereby generating the black pixel histogram in the vertical direction and the horizontal direction. Calculate for each horizontal direction. In addition, the setting unit 13 determines that a black pixel break is a point where the frequency included in the black pixel histogram is zero or less than a certain value. In the example of FIG. 2, in the horizontal direction, seven of n1, n2, n3, n4, n5, n6 and n7 are detected as black pixel breaks in the horizontal direction, and m1 and m2 in the vertical direction. Are detected as black pixel breaks in the vertical direction. For example, if the width of the rectangle for which the range has been specified does not match the horizontal black pixel break, the width direction edge that forms the rectangle at the black pixel break closest to the rectangle in the horizontal distance The length can be corrected. Also, if the height of the rectangle for which the range is specified does not match the cut in the black pixel in the vertical direction, the side in the height direction that forms the rectangle at the black pixel cut closest to the rectangle Can be corrected.

  As described above, the setting unit 13 is activated on the condition that the range is specified, and executes processing in the background so that a rectangular side near the position of the cut coincides with the position of the cut. Can do.

  The rectangular storage unit 14 is a storage unit that stores rectangular definition data.

  As an embodiment, the rectangular data set on the question sentence image by the setting unit 13 is stored in the rectangular storage unit 14. For example, it is assumed that the position and size of the rectangle are defined by the coordinates of the upper left vertex and the coordinates of the lower right vertex, but the rectangle may be defined by other parameters. Here, as an example, it is assumed that rectangular data is stored in a list format, but may be stored in another format, for example, a table format.

  The acquisition unit 15 is a processing unit that acquires handwriting data regarding the answer to the side line problem.

  As an embodiment, the acquisition unit 15 can also acquire by reading handwriting data stored in an auxiliary storage device such as a hard disk or an optical disk or a removable medium such as a memory card or a USB (Universal Serial Bus) memory. As another example, the acquisition unit 15 can also acquire handwriting data by receiving it from an external device, for example, a terminal device used by an answerer via a network.

  Such handwriting data is generated by handwriting input at the answerer's terminal device. For example, the answerer's terminal device accepts handwritten input drawn via a pointing device on the question sentence image displayed on the display unit of the terminal device. Thereby, time-series data of handwritten input coordinates on the problem sentence image, that is, a set of handwriting points is collected as handwriting data. For example, in handwriting data, coordinates of handwritten input are taken in by basic software such as an operating system at a sampling period of tens of seconds.

  The handwriting storage unit 16 is a storage unit that stores handwriting data.

  As one embodiment, the handwriting storage unit 16 stores handwriting data acquired by the acquisition unit 15, that is, time-series data of handwritten input coordinates. For example, the handwriting storage unit 16 uses a series of strokes detected from when a handwritten input is detected by a touch panel or the like until the handwritten input is no longer detected, so-called one stroke writing as a minimum unit of the list, and the handwriting data Are stored in list form. Although it is assumed here that the handwriting data is stored in a list format, it may be stored in another format, for example, a table format.

  The calculation unit 17 is a processing unit that calculates a determination range to be compared with handwriting data from a rectangle. In the following, a case where a side line is drawn in a horizontally written question sentence will be described as an example. However, the same determination is made by replacing the horizontal problem sentence vertically with the vertical sentence and the vertical sentence horizontally. Needless to say, you can.

  As an embodiment, the calculation unit 17 calculates the height h of the rectangle R using the rectangle data from the rectangle list stored in the rectangle storage unit 14. For example, since the rectangle data includes the coordinates of the upper left vertex and the coordinates of the lower right vertex with respect to the rectangle R, the Y coordinate of the lower right vertex is subtracted from the Y coordinate of the upper left vertex, or each other's Y The height h of the rectangle R can be obtained by calculating the difference in coordinates and taking the absolute value thereof.

  In addition, the calculation unit 17 uses the first determination to determine whether the deviation between the position of the side line determined by the handwriting data and the position of the correct character string is within the allowable range as the determination range. A threshold value t2 and a threshold value t3 are calculated.

  For example, when a side line is drawn in a horizontally written question sentence, a situation where an underline is drawn in a correct character string is assumed. Since the underline is drawn by handwriting input, it is difficult to avoid the underline from moving up and down from the correct character string, so there is a condition that the position of the underline is shifted from the correct character string up and down. Admit in. As an example of such a condition, a range in which a side line determined by handwriting data and a line in which a correct character string is described is set as the allowable range. That is, the threshold value t2 is a threshold value that allows the side line to shift upward from the lower end of the correct character string. For example, the height h of the rectangle R that can estimate the size of the character that forms the correct character string Is multiplied by a predetermined coefficient a2, for example, a value of 1 or less. Further, the threshold value t3 is a threshold value that allows a side line to shift downward from the lower end of the correct character string. For example, the height h of the rectangle R that can estimate the size of the character that forms the correct character string. Is multiplied by a predetermined coefficient a3, for example, a value of 1 or less. When the underline is drawn in this way, the underline is drawn between the lines. Therefore, it can be estimated that there is a higher possibility that the underline is shifted in the lower direction not overlapping the character string than in the upper direction overlapping the character string. From this, the threshold value t2 and the threshold value t3 can be the same value, but the coefficient a2 and the coefficient a3 that satisfy t2 <t3 can also be used for setting the threshold value.

  Further, the calculation unit 17 calculates a threshold value t1 used for the second determination for determining whether or not the deviation between the end point of the side line determined by the handwriting data and the end portion of the correct character string is within the allowable range.

  That is, when the underline is drawn by handwriting input in a horizontally written question sentence, it is difficult to avoid the underline being shifted to the left and right from the correct character string, so that each end point of the underline starts from both ends of the correct character string. A lateral deviation, such as protruding or not reaching both ends, is allowed with conditions. As an example of such a condition, a range in which an end point of a side line determined by handwriting data and a character at both ends of a correct character string can be associated is set as the above-described allowable range. That is, the threshold value t1 is a threshold value that allows the end points of the side lines to deviate in the horizontal direction from both ends of the correct character string. For example, the height of the rectangle R that can estimate the size of the character that forms the correct character string The value h can be set by multiplying a predetermined coefficient a1, for example, a value of 1 or less.

  The determination unit 18 is a processing unit that determines whether or not the handwriting is within a determination range determined from the frame. As illustrated in FIG. 1, the determination unit 18 includes a first determination unit 18a, a second determination unit 18b, and a third determination unit 18c.

  The first determination unit 18a is a processing unit that performs the first determination.

As one embodiment, the first determination unit 18a calculates an allowable range of vertical displacement with respect to the lower end position Y _low of the rectangle R. For example, the first determination unit 18a subtracts the threshold value t2 calculated by the calculation unit 17 from the lower end position Y _low obtained from the rectangle data of the rectangle R stored in the rectangle storage unit 14, that is, the calculation of Y _low −t2. Thus, the allowable range Y1 of the upward shift is calculated. Further, the first determination unit 18a performs a downward operation by adding the threshold value t3 calculated by the calculation unit 17 to the lower end position Y _low obtained from the rectangular data stored in the rectangular storage unit 14, that is, by calculating Y _low + t3. An allowable range Y2 of deviation is calculated. On the other hand, the first determination unit 18a calculates a circumscribed rectangle r of the handwriting from the handwriting data of the handwriting to be determined in the handwriting list stored in the handwriting storage unit 16. Further, the first determination unit 18a calculates the center position Yc in the vertical direction of the circumscribed rectangle r of the handwriting. Then, the first determination unit 18a determines whether the center position Yc of the circumscribed rectangle r of the handwriting is within the allowable range, that is, whether the condition “Y1 ≦ Yc ≦ Y2” is satisfied.

  When this condition is met, even if the position of the side line is shifted up and down from the bottom of the correct character string, the underline is in the vertical direction up to the next line up to or below the correct character string. It can be estimated that there is no deviation. On the other hand, when this condition is not satisfied, it can be estimated that the underline is shifted to the line immediately above or to the line below by 1 in the correct character string.

  The second determination unit 18b is a processing unit that performs the second determination.

As an embodiment, the second determination unit 18b calculates a right end position X _R of the leftmost position X _L and rectangle R of the rectangle R. For example, the second determination unit 18b extracts a upper left X coordinate of the vertex included in the rectangle data of the stored rectangle R in a rectangular storage section 14 as the left end position X _L of the rectangle R, for the lower right apex extracting the X-coordinate as a position X _R of the left edge of the rectangle R. At the same time, the second determination unit 18b extracts the _left end position X _left of the circumscribed rectangle r of the handwriting obtained by the first determination unit 18a, and extracts the right end position X _right of the circumscribed rectangle r. On top of that, the second determination unit 18b is whether the position X _left of the left edge of the circumscribed rectangle r is within the allowable range, i.e. whether or not the condition _{"X L -t1 ≦ X left ≦ X} L + t1 " Determine. Then, the second determination unit 18b determines whether or not the _right end position X _right of the circumscribed rectangle r is within the allowable range, that is, whether or not the condition “X _R −t1 ≦ X _right ≦ X _R + t1” is satisfied. To do.

  When these two conditions are met, even if each end point of the underline protrudes from both ends of the correct character string or does not reach both ends, it is only underlined to the extent that the characters at both ends of the correct character string can be associated with each other. It can be estimated that the end points of are not displaced in the left-right direction. On the other hand, if any one of the two conditions is not satisfied, an underline is drawn up to a character that is not included in the correct character string, or an underline is not drawn in all the correct character strings. It can be estimated that there is excess or deficiency. Here, the case where the determination is performed on both the left and right ends of the circumscribed rectangle r is illustrated, but only one of the determinations may be performed.

  The 3rd determination part 18c is a process part which performs the 3rd determination which determines whether the handwriting determined by handwriting data is linear form.

  As an embodiment, the third determination unit 18c can employ determination logic that determines whether or not at least the shape of the handwriting is a straight line, and further determines the direction of the handwriting. For example, if the question sentence is horizontal writing, it is determined whether or not it is a straight line extending in the horizontal direction, and if it is vertical writing, it is determined whether it is a straight line extending in the vertical direction. As an example of a method for determining whether or not the shape is a straight line, a method using a least square method can be given. Specifically, the third determination unit 18c determines that there is an error between the straight line that best represents the handwriting sequence that forms the handwriting, that is, the time series data of the coordinates included in the handwriting data when the question sentence is horizontal writing. The minimum horizontal straight line is obtained by the least square method, and the positional deviation between this straight line and the stroke array, in this example, the vertical deviation is calculated. Then, the third determination unit 18c determines whether or not the shape of the handwriting is linear depending on whether or not a statistical value of positional deviation, for example, an average value is within a certain value. In addition, in the handwriting data, the third determination unit 18c performs a vector between the brush points adjacent in time series, calculates the average of the deviation between the direction of each vector and the predetermined direction, and calculates the average value of the deviation. It is also possible to determine whether or not the shape of the handwriting is linear depending on whether or not is within a predetermined range. Furthermore, simply, a method using the ratio between the height and width of the circumscribed rectangle r of the handwriting is also effective. For example, when the height h and the width w are in a relationship of w> h × 3, the handwriting can be regarded as a straight line extending approximately in the lateral direction. Note that any of the above methods can be adopted alternatively or combined, and it can be determined whether or not the handwriting is linear by other existing techniques.

  The scoring unit 19 is a processing unit that scores the answer to the side line problem according to the determination result of the determination unit 18.

  As one embodiment, the scoring unit 19 gives a correct answer mark to a combination of a rectangle and a handwriting determined to satisfy the conditions in all the processing units of the first determination unit 18a, the second determination unit 18b, and the third determination unit 18c. To do. On the other hand, the scoring unit 19 includes the first determination unit 18a and the first determination unit 18a for all combinations of rectangles included in the rectangle list stored in the rectangle storage unit 14 and handwriting included in the handwriting list stored in the handwriting storage unit 16. After the determination by the 2 determination unit 18b and the third determination unit 18c, an incorrect answer mark is assigned to a rectangle and a handwriting to which no correct answer mark is assigned. By assigning such correct answer marks and incorrect answer marks, it is possible to evaluate the score and the like. For example, the number of correct marks can be totaled according to a predetermined score for each rectangle, or the ratio of correct marks to the number of rectangles can be calculated.

  The function units such as the setting unit 13, the acquisition unit 15, the calculation unit 17, the determination unit 18, and the scoring unit 19 can be implemented as follows. For example, it can be realized by causing a central processing unit, a so-called CPU (Central Processing Unit), or the like to develop and execute a process that exhibits the same function as each of the above-described processing units on a memory. These processing units do not necessarily have to be executed by the central processing unit, but may be executed by an MPU (Micro Processing Unit). Each functional unit described above can also be realized by a hard wired logic such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

  In addition, for example, various semiconductor memory elements such as RAM (Random Access Memory) and flash memory can be employed for the rectangular storage unit 14 and the handwriting storage unit 16. Further, the rectangular storage unit 14 and the handwriting storage unit 16 are not necessarily the main storage device, and may be an auxiliary storage device. In this case, an HDD (Hard Disk Drive), an optical disk, an SSD (Solid State Drive), or the like can be employed.

[Concrete example]
Next, a specific example of scoring the answer to the side line problem will be described with reference to FIGS. 3A to 3D. FIG. 3A is a diagram illustrating a setting example of a rectangle in a question sentence image. FIG. 3B is a diagram illustrating an example of entering an answer to the question sentence image. FIG. 3C is a diagram illustrating an example of a first determination method. FIG. 3D is a diagram illustrating an example of a second determination method. In these examples of FIGS. 3A to 3D, as an example of the sideline problem, the question sentence “I played in the schoolyard yesterday” is illustrated, and the correct answer string “in the schoolyard” is underlined. Assume a case of determining whether or not. In addition, although the case where a problem sentence is horizontal writing is illustrated here, in the case of vertical writing, it may be determined whether or not a side line is drawn on the right side of the “in the schoolyard” portion.

  As shown in FIG. 3A, the question sentence is given as an image, and the question sentence image describes “I played in the schoolyard yesterday”. However, since it is handled as an image, the computer cannot recognize what is written as text data. In this question sentence image, a rectangle R is set by specifying a range such as a drag operation at a portion of “in the schoolyard” which is a correct character string. The rectangle is set by the problem creator via the touch panel 11 or the like. As a result, a sideline problem is created.

  FIG. 3B shows a state in which the answer is written on the question sentence image by the answerer. In the example shown in FIG. 3B, an underline U is drawn at a portion “in the schoolyard” by handwriting input. When the handwritten data of the underline U is acquired from the answerer's terminal device, the first determination is performed as shown in FIG. 3C.

As shown in FIG. 3C, an allowable range in which the distance from the lower end position Y _low of the rectangle R to the upper direction is set as the threshold value t2 and the downward distance is set as the threshold value t3, that is, the filled region shown in the figure is set. Whether or not the center position Yc of the circumscribed rectangle r of the underline U is included is determined. In this example, the center position Yc of the circumscribed rectangle r of the underline U is included in the allowable range. Therefore, it can be estimated that the underline is not shifted in the vertical direction up to the line immediately above or the line below the correct character string. Here, the case where it is determined whether or not the center position Yc of the circumscribed rectangle r of the underline U is included in the allowable range is illustrated, but as a determination condition that both the upper end and the lower end are included in the allowable range Also good.

Thus, when satisfy | filling conditions by 1st determination, 2nd determination is performed in the way shown to FIG. 3D. As shown in FIG. 3D, it is determined whether or not the _left end position X _left and the right end position X _right of the underline U fall within the distance t1 from the position of the left and right ends of the rectangle R, respectively. In this example, since the left end position X _left and the right end position X _right of the underline U are accommodated, the underline U can only be correlated with the characters “school” and “de” at both ends of the correct character string. It can be estimated that each end point is not shifted in the left-right direction. Thereafter, when it is determined that the shape of the underline U is a straight line, that is, a horizontal straight line, a correct answer mark is given to the rectangle R and the underline U. Here, the distance t1 is set to the left and right from the end of the rectangle R, but different values calculated based on the height h may be set.

  As described above, the first determination, the second determination, and the third determination are answers by handwriting input even if character recognition processing such as OCR is not performed by setting a rectangle and acquiring handwriting data. It can be determined whether or not is a correct answer or an incorrect answer. Therefore, since the correction operation for the character recognition result can be made unnecessary, it is possible to automate the association between the correct character string of the question sentence and the character string of the side line portion. Therefore, the scoring efficiency of the side line problem can be improved.

[Process flow]
4 and 5 are flowcharts illustrating the procedure of the marginal line problem scoring process according to the first embodiment. This process is executed when a question sentence to be scored is specified by a person in charge of scoring, for example, a question creator, with the vertical or horizontal writing type of the question sentence set.

  As shown in FIG. 4, the reference position referring to the rectangular list stored in the rectangular storage unit 14 is initialized to the top (step S101). Subsequently, the following steps S103 to S119 are executed until the reference position of the rectangular list is set at the end (No in step S102).

  That is, the calculation unit 17 selects rectangular data corresponding to the reference position being set from the rectangular list (step S103), and shifts the reference position of the rectangular list to the next rectangular data (step S104). Subsequently, the calculation unit 17 calculates the height h of the rectangle R (step S105).

  Then, the calculation unit 17 uses the height h of the rectangle R calculated in step S105, the threshold value t2 and the threshold value t3 that allow the side line to deviate vertically from the lower end of the correct character string, and the correct character A threshold t1 that allows the end point of the side line to shift from the both ends of the column in the left-right direction is calculated as an allowable range, that is, a determination range (step S106).

Subsequently, the first determination unit 18a performs the calculation of subtracting the threshold value t2 calculated in step S106 from the lower end position Y _low obtained from the rectangular data of the rectangle R selected in step S103, thereby allowing an upward shift tolerance. By calculating the range Y1 and adding the threshold value t3 calculated in step S106 to the lower end position Y _low obtained from the rectangular data of the rectangle R selected in step S103, an allowable range Y2 of downward shift is obtained. Is calculated (step S107).

  After that, the reference position referring to the handwriting list stored in the handwriting storage unit 16 is initialized to the top (step S108), and the reference position of the handwriting list is set to the end (No in step S109). Then, the following steps S110 to S119 are executed.

  That is, the first determination unit 18a selects handwriting data corresponding to the reference position being set from the handwriting list (step S110), and shifts the reference position of the handwriting list to the next handwriting data (step S111). Subsequently, the first determination unit 18a calculates a circumscribed rectangle r of the handwriting from the handwriting data selected in Step S110 (Step S112). Then, the first determination unit 18a calculates the center position Yc of the circumscribed rectangle r of the handwriting (Step S113).

  Then, the first determination unit 18a determines whether the center position Yc of the circumscribed rectangle r of the handwriting is within the allowable range, that is, whether the condition of “Y1 ≦ Yc ≦ Y2” is satisfied (step S1). S114).

At this time, when the center position Yc of the circumscribing rectangle r of the handwriting satisfies the condition of “Y1 ≦ Yc ≦ Y2” (Yes at Step S114), the second determination unit 18b is included in the rectangular data of the rectangle R selected at Step S103. the X coordinate of the upper left vertex to extract the X-coordinate of the lower right vertex are extracted as the left end position X _L of the rectangle R as the right end position X _R of the rectangle R being circumscribed rectangle of handwriting obtained in step S112 The left end position X _left of r is extracted and the _right end position X _right of the circumscribed rectangle r is extracted (step S115).

On top of that, the second determination unit 18b is whether the position X _left of the left edge of the circumscribed rectangle r is within the allowable range, i.e. whether or not the condition _{"X L -t1 ≦ X left ≦ X} L + t1 " Is determined (step S116). At this time, when the position X _left of the left edge of the circumscribed rectangle r satisfies the condition _{"X L -t1 ≦ X left ≦ X} L + t1 " (step S116Yes), the second determination unit 18b, the position of the right edge of the circumscribed rectangle r It is further determined whether or not X _right is within an allowable range, that is, whether or not the condition of “X _R −t1 ≦ X _right ≦ X _R + t1” is satisfied (step S117).

If the right end position X _right of the circumscribed rectangle r also satisfies the condition “X _R −t1 ≦ X _right ≦ X _R + t1” (step S117 Yes), the third determination unit 18c selects the handwriting selected in step S110. It is determined whether or not the handwriting determined by the data is linear (step S118).

  When the handwriting is linear (step S118 Yes), the scoring unit 19 gives a correct answer mark to the combination of the rectangle selected in step S103 and the handwriting selected in step S110 (step S119). The process proceeds to S109.

  On the other hand, if any of the above steps S114No, S116No, S117No, or S118No is satisfied, the process proceeds to step S109 without giving a correct answer mark.

  After that, when the reference position of the handwriting list is the tail and the reference position of the rectangle list is the tail (Yes in Step S109 and Step S102), the scoring unit 19 uses the rectangle and the correct mark not assigned in Step S119. An incorrect answer mark is given to the handwriting (step S120), and the process ends.

[One aspect of effect]
As described above, the sideline problem scoring device 10 according to the present embodiment calculates a determination range based on a rectangle set in a region including a character string that is a correct answer on an image including a question sentence, and is based on handwriting input. It is determined whether or not the handwriting falls within the determination range. Therefore, according to the side line problem scoring device 10 according to the present embodiment, the scoring efficiency of the side line problem can be improved.

  Although the embodiments related to the disclosed apparatus have been described above, the present invention may be implemented in various different forms other than the above-described embodiments. Therefore, another embodiment included in the present invention will be described below.

[Application example]
When the distance between the left or right side of the first rectangle and the left or right side of the second rectangle different from the first rectangle is within a predetermined distance, The one rectangle and the second rectangle can be connected to the left and right.

  FIG. 6 is a diagram illustrating an application example. FIG. 6 shows a first rectangle R1 and a second rectangle R2 that are connected to each other. As shown in FIG. 6, the distance between the right side of the first rectangle R1 and the left side of the second rectangle R2 is determined based on the heights of the first rectangle R1 and the second rectangle R2. If it is equal to or less than the threshold value, the first rectangle R1 and the second rectangle R2 are connected to be combined into one rectangle, and the combined rectangle is added to the rectangle list. At this time, the original first rectangle R1 and the original second rectangle R2 are not deleted from the rectangle list.

  “H” shown in FIG. 6 represents the height of the rectangle. Here, “h” may be obtained by taking the average of the heights of a plurality of rectangles, or one of the heights of any rectangle may be selected as a representative to set the value of h. The distance threshold value TW and the threshold value TH are obtained by multiplying the “h” by a fixed number AW and AH, respectively. Also, let dW be the distance on the horizontal axis between the nearest sides of the two rectangles on the left and right sides. If the position of the straight line passing through the center of the rectangle in the vertical direction is the position in the vertical direction of the rectangle, the distance value in the vertical direction of the two rectangles is dH as shown in FIG. When the horizontal distance dW is equal to or smaller than the threshold TW and the vertical distance dH is equal to or smaller than the threshold TH, the two rectangles are merged. The rectangle data obtained by merging these two rectangles is added to the rectangle list, and the processes shown in FIGS. 4 and 5 are executed. In the example of FIG. 6, the case where the question sentence is written horizontally is illustrated, but the calculation can be similarly applied to the case of vertical writing by switching the vertical and horizontal directions in the calculation process.

  Thereby, both when the underline is drawn in each of the first rectangle R1 and the second rectangle R2 and when the underline is drawn together in the first rectangle R1 and the second rectangle R2, The correct answer can be determined.

  7A and 7B are diagrams illustrating one aspect of the effect of the application example. In both FIG. 7A and FIG. 7B, the question sentence is “acquired within a specific range until now”, and the two character strings “specific” and “within range” are correct. It is shown. FIG. 7A shows an answer when an underline U1 is drawn at a position corresponding to the correct character string “specific” and an underline U2 is drawn at a position corresponding to the correct character string “within range”. It is shown. On the other hand, FIG. 7B shows a case where the underline U3 is drawn over the entire character string “within a specific range” including two correct character strings.

  There may be a case where both the underline U1 and the underline U2 shown in FIG. 7A and the underline U3 shown in FIG. This is because there is a possibility that the underline that the answerer intends to underline the character string “within a specific range” or that the underline drawn unintentionally is connected to each rectangle. Even in such a case, an underline U1 and an underline U2 shown in FIG. 7A and an underline U3 shown in FIG. 7B are added to the rectangle list by adding a rectangle in which the first rectangle R1 and the second rectangle R2 are combined. Both can be determined as correct answers. 7A and 7B exemplify the case of horizontal writing, but in the case of vertical writing, the upper or lower side of the first rectangle and the upper side of the second rectangle different from the first rectangle or When the distance to the lower side is within the range of values calculated based on the height or width of the rectangle, the same effect can be obtained by connecting the first rectangle and the second rectangle vertically. be able to.

[Vertical writing]
In the first embodiment described above, the case where the question sentence is written horizontally is illustrated as an example. However, even when the question sentence is written vertically, the processes shown in FIGS. 4 and 5 can be executed in the same manner. For example, in the case of the first determination, the horizontal position stored in the rectangle storage unit 14 is compared with the horizontal position of the handwriting stored in the handwriting storage unit 16, and the difference in position is obtained based on the width of the rectangle. What is necessary is just to determine whether it is in the tolerance | permissible_range. Further, in the case of the second determination, the positions of the upper and lower ends of each rectangle stored in the rectangle storage unit 14 are compared with the positions of the upper and lower ends of the handwriting stored in the handwriting storage unit 16, and the difference in position is rectangular. What is necessary is just to determine whether it is in the tolerance | permissible_range calculated | required based on the width | variety of.

[Automatic layout detection]
In the first embodiment, the case where the question sentence is set to horizontal writing or vertical writing is exemplified. However, the question sentence is written horizontally or vertically from the aspect ratio of each rectangle stored in the rectangle storage unit 14. It can be set automatically. For example, when the horizontal side of a rectangle is longer than the vertical side, it can be estimated as horizontal writing, and when the vertical side of the rectangle is longer than the horizontal side, horizontal writing is performed. Can be estimated. In this case, obtaining the same determination result with a plurality of rectangles may be a condition for confirming the determination of horizontal writing or vertical writing.

[Distribution and integration]
In addition, each component of each illustrated apparatus does not necessarily have to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. For example, the setting unit 13, the acquisition unit 15, the calculation unit 17, the determination unit 18, or the scoring unit 19 may be connected as an external device of the side line problem scoring device 10 via a network. Moreover, another apparatus has the setting part 13, the acquisition part 15, the calculation part 17, the determination part 18, or the scoring part 19, respectively, and the function of said sideline problem scoring apparatus 10 is connected by network connection and cooperating. It may be realized.

[Side Line Problem Scoring Program]
The various processes described in the above embodiments can be realized by executing a prepared program on a computer such as a personal computer or a workstation. Therefore, in the following, an example of a computer that executes a sideline problem scoring program having the same function as that of the above embodiment will be described with reference to FIG.

  FIG. 8 is a diagram illustrating a hardware configuration example of a computer that executes the sideline problem scoring program according to the first embodiment and the second embodiment. As illustrated in FIG. 8, the computer 100 includes an operation unit 110a, a speaker 110b, a camera 110c, a display 120, and a communication unit 130. Further, the computer 100 includes a CPU 150, a ROM 160, an HDD 170, and a RAM 180. These units 110 to 180 are connected via a bus 140.

  As shown in FIG. 8, the HDD 170 has a sideline problem scoring program 170a that exhibits the same functions as the setting unit 13, the acquisition unit 15, the calculation unit 17, the determination unit 18, and the scoring unit 19 shown in the first embodiment. Is memorized. This side line problem scoring program 170a may be integrated or separated as with the constituent elements of the setting unit 13, the acquisition unit 15, the calculation unit 17, the determination unit 18, and the scoring unit 19 shown in FIG. That is, the HDD 170 does not necessarily have to store all the data shown in the first embodiment, and data used for processing may be stored in the HDD 170.

  Under such an environment, the CPU 150 reads the side line problem scoring program 170 a from the HDD 170 and develops it in the RAM 180. As a result, the side line problem scoring program 170a functions as a side line problem scoring process 180a as shown in FIG. The sideline problem scoring process 180a expands various data read from the HDD 170 in an area allocated to the sideline problem scoring process 180a in the storage area of the RAM 180, and executes various processes using the expanded various data. . For example, the process shown in FIG. 4 and FIG. 5 is included as an example of the process executed by the side line problem scoring process 180a. Note that the CPU 150 does not necessarily operate all the processing units described in the first embodiment, and the processing unit corresponding to the process to be executed may be virtually realized.

  The sideline problem scoring program 170a may not necessarily be stored in the HDD 170 or the ROM 160 from the beginning. For example, each program is stored in a “portable physical medium” such as a flexible disk inserted into the computer 100, so-called FD, CD-ROM, DVD disk, magneto-optical disk, or IC card. Then, the computer 100 may acquire and execute each program from these portable physical media. In addition, each program is stored in another computer or server device connected to the computer 100 via a public line, the Internet, a LAN, a WAN, etc., and the computer 100 acquires and executes each program from these. It may be.

DESCRIPTION OF SYMBOLS 10 Side line problem scoring apparatus 11 Touch panel 12 Problem memory | storage part 13 Setting part 14 Rectangular memory | storage part 15 Acquisition part 16 Handwriting memory | storage part 17 Calculation part 18 Judgment part 18a 1st judgment part 18b 2nd judgment part 18c 3rd judgment part 19 Scoring part

Claims (7)

Computer
Get the frame that is set to the correct text string of the question text included in the image,
Calculate the allowable range that allows deviation from the set frame,
Get handwriting by handwriting input,
It is determined whether or not the sideline problem corresponding to the frame is a correct answer by whether or not there is a handwriting included within the allowable range calculated from the frame ,
A sideline problem scoring method characterized by executing processing. The calculating process calculates the allowable range based on the width or height of the frame,
The determining process, in the handwriting by previous SL handwriting input, depending on whether the handwriting position of the right and left ends or the upper and lower ends on the image is included within the allowable range is present, corresponding to the frame underline The side line problem scoring method according to claim 1, wherein it is determined whether or not the problem is a correct answer . The calculating process calculates the allowable range based on the width or height of the frame,
The determining process, in the handwriting by previous SL handwriting input, depending on whether the handwriting vertical or horizontal position on the image is included within the allowable range is present, the underline problem corresponding to the frame The side line problem scoring method according to claim 1, wherein it is determined whether or not the answer is correct . The computer is
The process of determining whether or not the handwriting input by the handwriting input is linear when the handwriting is within the calculated allowable range is further performed. The sideline problem scoring method described in 1. The computer is
When the distance between the left and right or upper and lower sides of the first frame and the left and right or upper and lower sides of the second frame different from the first frame is within a range determined from the height or width of the frame , Further executing a process of combining the first frame and the second frame by connecting left and right or up and down,
The determination process is characterized in that for each of the first frame, the second frame, and the combined frame, it is determined whether there is a handwriting that falls within an allowable range calculated from the frame. The side line problem scoring method according to any one of claims 1 to 4. On the computer,
Get the frame that is set to the correct text string of the question text included in the image,
Calculate the allowable range that allows deviation from the set frame,
Get handwriting by handwriting input,
It is determined whether or not the sideline problem corresponding to the frame is a correct answer by whether or not there is a handwriting included within the allowable range calculated from the frame ,
A sideline problem scoring program characterized by executing processing. A first acquisition unit that acquires a frame set to a correct character string among character strings of question sentences included in the image;
A calculation unit that calculates an allowable range that allows deviation from the set frame;
A second acquisition unit for acquiring handwriting by handwriting input;
A determination unit that determines whether or not the sideline problem corresponding to the frame is a correct answer, depending on whether or not the handwriting included in the allowable range calculated from the frame exists;
A side line problem scoring device characterized by comprising:

Images